Radionuclides Deposition Over Antarctica M

Journal of Environmental Radioactivity 68 (2003) 137–158 www.elsevier.com/locate/jenvrad

Radionuclides deposition over Antarctica M. Pourchet a, O. Magand a,∗, M. Frezzotti b, A. Ekaykin c, J.-G. Winther d

a Laboratoire de Glaciologie et Ge´ophysique de l’Environnement, CNRS, BP 96, 38402 St Martin d’He`res Cedex, France b ENEA, CLIM-OSS, P.O. Box 2400, 00100 Rome A.D., Italy c Arctic and Antarctic Research Inst., 38 Bering St., 199397 St. Petersburg, Russia d Norwegian Polar Institute, Polar Environmental Centre, Tromso, Norway

Received 17 June 2002; received in revised form 11 February 2003; accepted 21 February 2003

Abstract

A detailed and comprehensive map of the distribution patterns for both natural and artificial radionuclides over Antarctica has been established. This work integrates the results of several decades of international programs focusing on the analysis of natural and artificial radionuclides in snow and ice cores from this polar region. The mean value (37±20 Bq mϪ2)of 241Pu total deposition over 28 stations is determined from the gamma emissions of its daughter 241Am, presenting a long half-life (432.7 yrs). Detailed profiles and distributions of 241Pu in ice cores make it possible to clearly distinguish between the atmospheric thermonuclear tests of the fifties and sixties. Strong relationships are also found between radionuclide data (137Cs with respect to 241Pu and 210Pb with respect to 137Cs), make it possible to estimate the total deposition or natural fluxes of these radionuclides. Total deposition of 137Cs over Antarctica is estimated at 760 TBq, based on results from the 90–180° East sector. Given the irregular distribution of sampling sites, more ice cores and snow samples must be analyzed in other sectors of Antarctica to check the validity of this figure.  2003 Elsevier Science Ltd. All rights reserved.

Keywords: Radionuclides deposition; Flux; Antarctica

∗ Corresponding author. Tel.: +33-4-76-82-42-59; fax: +33-4-76-82-42-01. E-mail address: [email protected] (O. Magand).

1. Introduction

Artificial radioisotopes resulting from atmospheric thermonuclear tests carried out between 1953 and 1980 were deposited in Antarctica after transport in the upper atmosphere and stratosphere, creating distinct radioactive reference levels in the snow. The dates of arrival and deposition in this polar region are well known and therefore provide a means to estimate Antarctic snow accumulation rates or describe air mass circulation patterns (Wilgain et al., 1965; Feely et al., 1966; Pourchet et al., 1983; Pourchet et al., 1997). The 1955 and 1965 radioactivity peaks provide two very convenient horizons for dating snow and ice layers and measuring accumulation. Special techniques have been developed over the last 40 yrs to detect and measure artificial and natural radionuclides present in the ice sheets (Picciotto and Wilgain, 1963; Delmas and Pourchet, 1977; Pinglot and Pourchet, 1979, 1994). In Antarctica, total beta counts remain the most frequent radioactivity measurement and are used to determine the average rate of snow accumulation (Picciotto and Wilgain, 1963; Lambert et al., 1977). Gamma profiles are also used for the determination of artificial and natural radionuclide fluxes in the region. The total deposition of artificial radionuclides (mainly 137Cs, 90Sr and 3H) and the flux of natural radionuclides (especially 210Pb) has been estimated using snow samples from many Antarctic study areas (Pourchet et al., 1997), although simultaneous measurements of these major and complementary radionuclides in the same snow core are very rare. Snow and ice sheets on the Antarctic continent are not the only depositories of artificial and natural radionuclides deposited from the atmosphere over the south polar region. The deposition and flux of these radionuclides have also been measured in soils, plants and lake sediments, for example on the Antarctic Peninsula (Roos et al., 1994). A detailed profile of certain radionuclides has also been obtained for the Ross Ice Shelf (Koide et al., 1979). In this work we have extended the previously available total deposition data by adding the results of a simultaneous analysis of 137Cs, 210Pb and 241Pu in the same snow cores for about 30 stations located in various parts of Antarctica and a detailed profile of these three radionuclides for Vostok and Dome Concordia stations on the Antarctic plateau area. Key results of several decades of international programs and experiments focusing on artificial radionuclide analyses of snow or ice core samples in Antarctica are used in this paper. In a previous paper (Pourchet et al., 1997), a distribution of artificial and natural radionuclides over Antarctica was proposed, based on 137Cs, 210Pb and 3H data from snow and ice-core samples. In this work, additional radioactive analysis data are used to assess and extend the previous data, increasing the significance of statistical results and validating the distribution pattern previously proposed. The evidence of a strong correlation between radionuclides may make it possible to estimate the deposition and flux of one radionuclide from available data for another one. For 210Pb, field data assess the possible double origin (unsupported and supported 210Pb) of this terrigenous radionuclide in Antarctic snow and ice-core samples, a subject already discussed by Pourchet et al. (1997). Finally, the establishment of a distribution pattern for artificial and natural radionuclides, which constitute isotopic tra- M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137–158 139 cers, can help us to understand the arrival and deposition of other artificial and natural tracers via atmospheric transport.

2. Material and methods

2.1. Study area and sample description

The snow and ice samples described in this paper were collected from cores or pits in different regions of the Antarctic continent, mainly in the 90–180° E sector from the Russian (Mirny to Vostok), Australian (Casey to Vostok), French (Dumont d’Urville to Dome C) and Italian (Talos Dome to Dome Concordia) transects. Ice- coring stations from other programs close to these transects were also used. Twenty-eight stations previously analyzed for 137Cs (Pourchet et al., 1997) were subject to new simultaneous 137Cs, 210Pb and 241Pu measurements. The same analyses were also performed for 24 new locations (Table 1). For Vostok and Dome Concordia stations, detailed radionuclide deposition measurements were carried out (Table 2).

2.2. Analytical methods and procedure

Using a method developed by Delmas and Pourchet (1977), snow and ice core samples were melted, weighed, acidified and filtered on ion exchange paper, where all radionuclides were trapped. After drying, the filters were directly analyzed by gamma spectrometry using a low background germanium detector (Pinglot and Pourchet, 1994). Because of low activity, some samples were combined for radioactivity analysis (Pinglot and Pourchet, 1979, 1994). Most of the studied samples were analyzed at the LGGE laboratory in Grenoble. For high resolution gamma spectrometry, the analyzer was protected against all interfering ambient radioactivity, in particular using an anti-Compton device. This system provides a lower detection threshold, especially for the isotopes of interest such as 137Cs (30.2 yrs), 210Pb (22.3 yrs) and 241Am (432.7 yrs), daughter of 241Pu (14.4 yrs) (Pinglot and Pourchet, 1994, 1995; Pinglot et al., 1999). Standard liquid 137Cs, 210Pb and 241Am liquids from the CEA or Amersham labora- tories (2% uncertainty at 95% of confidence level) were used to calibrate the detector. The analytical procedures were the same as those used for the snow samples. The accuracies for our measurements are about 20% for 137Cs and 210Pb and 50% for 241Am. New determinations of 137Cs for the 28 previously analyzed stations (Pourchet et al., 1997) confirm the accuracy of previous results (Fig. 1). The specific activities have been corrected for decay to the deposition time.

2.3. Measurements and determination of 241Pu and 137Cs

The 241Pu and 137Cs radionuclides constitute two well-known debris products of atmospheric thermonuclear tests between 1955 and 1980. 137Cs is characteristic of both well-known reference layers in snow (1955 and 1965 peaks on beta profiles), 140 M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137–158 ) ) 1 Ϫ yr 2 Ϫ Pb Ref. continued on next page ( ux deposition fl 210 ) measured and calculated in 1 Ϫ yr Pu n.m. n.m. 3 ) 2 2 241 Ϫ Ϫ ∗ uxes (Bq m fl Cs Pb 137 210 ) (a) (b) (Bq m 1 Ϫ ), and 2 yr Ϫ 2 Ϫ 29.430.43337 9040 842953.9 69 140 15455.4 27155.4 1755.4 6455.4 89 18 23 16 19 9 23 4438.9 1.90 95 23 2.3643.6 52 81 18152.1 n.m. 3629.4 1.28 86 42 99 121 35 5.60 126 80 5.40 1 4.48 52 2 38 45 476 3.43 58 34 28 2 109 50 2 1.20 2.66 n.m. 49 55 2 2 n.m. 59 36 1.93 2 46 78 39 27 23 83 2 2 1.74 n.m. 3.52 2 n.m. 1.85 2 3.45 2 2 2 2 C) (kg m ° Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ temperature accumulation Pu density activities (Bq m 241 Cs and 137 )(m)( ° )( ° ( J9 82.37 191.33 50 Stations Lat. S Long.Ice E cores M.M Elevation 10 m 77.83 Snow 167.00 20 Deposition (Bq m 160 164 F9km105km260km325km400km200 67.58 84.4VK631 68.77Komso 69.3VK747 93.70 69.95 189.00VK14 94.47 68.25 1351VK17 72.22 95.02 2280 Vostok 50 95.62 74.1 (a)Vostok 73.25 94.08 2560 (b) 2760 DB 96.62 78.45 1970 BHD 78.45 97.50 78.45 3421BHDB 97.07 78.45GC38 106.84 3498 3270 106.84GC40 106.84 3470 106.84GC47 3471 76.92 3470 313 66.73GD03 66.73 3471 70.44 112.83 95.17 71.17 112.83 73.70 1315 3400 111.74 1315 69.00 68 111.36 51 2524 110.20 2695 84 155 115.48 3046 1832 51 52 4.22 56 40 631 631 32 18 48 2 29 2.55 59 174 2.26 81 153 59 67 40 2 n.m. n.m. 5.37 2 1.09 4.49 2 2 2 snow and ice-core samples of different zones of the Antarctic continent Table 1 Inventory of existing accumulation data, M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137–158 141 ) 1 Ϫ yr 2 Ϫ Pb Ref. ux deposition fl 210 Pu n.m.n.m. n.m.n.m. n.m.n.m. n.m.n.m. n.m.n.m. n.m.n.m. n.m. 3 n.m. n.m. 3 n.m. n.m. 3 n.m. n.m. 3 n.m. n.m. 3 n.m. 3 3 4 5 5 5 ) 2 241 Ϫ ∗ ∗ ∗ ∗ ∗ ∗ ∗ ∗ ∗ ∗ ∗ Cs 137 ) (a) (b) (Bq m 1 Ϫ yr 2 Ϫ 0.7 80 53 53 37 1.64 2 4 681 16 32.633.328.732.752.9 35013.4 35611.3 36750.7 293 65 163 95 236 100 103 80 63 107 46 n.m. 60 n.m. 62 32 3.77 7.51 40 n.m. 25 56 n.m. 3.63 45 2.33 19 30 67 n.m. 2 2 3.58 1.57 4.57 56 2 2 1.91 2 2 2 2 49.248.647.9 36 6110.714.3 551515.3 140 220 26 250 350 45 33 90 63 114 119 74 n.m. 6.32 2 C) (kg m Ϫ Ϫ ° Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ temperature accumulation )(m)( ° )( ° ( ) continued Stations Lat. S Long.GD12 E Elevation 10 m 68.98 Snow 126.93 2170 Deposition (Bq m GD15GF01GF04GM13JRF 69.00JRG 68.50PS (a) 68.53PS 130.81 (b) 73.17New 110.85 Byrd 2150 107.24 64.25 1800 110.45 64.25 90.00 2120 80.00 2960 90.00 302.25 302.25 1690 90.00 240.00 1350 90.00 2800 1500 2800 180 131 R. Boudoin2-11-2791-15-4151-13-370 70.431-10-275 82.902-0-0 85.20 24.32Livingston Is. 85.80D.Du 18.20 86.60A3 62.67 20 1.60D9 2610 82.10 8.70D12 30.60 2630 299.62 2690 66.70 2860 265 55.10 66.69 139.80 3720 -50 66.70 66.70 139.82 43 139.82 400 139.78 190 223 42 246 31 195 29 21 Table 1 ( 142 M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137–158 ) 1 Ϫ yr 2 Ϫ Pb Ref. ux deposition fl 210 Pu n.m. n.m.n.m. n.m.n.m. n.m. 6 n.m.n.m. n.m. 6 n.m.n.m.n.m. n.m. 6 n.m. 8 6 n.m. n.m. 5 5 7 ) 2 241 Ϫ ∗ ∗ ∗∗ ∗ ∗ ∗ ∗ ∗ Cs 137 ) (a) (b) (Bq m 1 Ϫ yr 2 Ϫ 17.517.918.32222.6 44024 275 139 320 240 420 44 26 88 54 65 39 52 42 n.m. 74 50 55 n.m. 5.55 57 158 28 4.04 4.83 7.28 4 2 2 2 2 25.829.8 26036.3 36041.241.4 260 7852.353.5 230 73 240 50 125 8025.6 32 4.55 n.m. 89 n.m. 92 135 59 95 n.m. n.m. 2 36 32 4 5.51 43 22 2 60 1.13 2 n.m. n.m. 2 9, 10 27.631.134.3 24038.8 70 32046.3 20053.5 130 20 64 1928.1 113 32 n.m. 113 0.82 171 52 2 47 16 44 1.07 n.m. n.m. 2 9, 10 C) (kg m ° Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ temperature accumulation )(m)( ° )( ° ( ) continued Stations Lat. S Long.D23 E Elevation 10 m 66.73 Snow 139.62 556 Deposition (Bq m D26D33D42D43D45 66.77 66.80 66.99 139.55 67.06 139.43 67.22 139.11 589 139.01 690 1028 138.83 1143 1353 D47D53D60 67.39D72 67.86 138.64D80 1524 137.99D120 68.45DC (a) 1898 69.44 137.20 70.02 2291 135.76 73.06EPICA A 74.64 134.82 2412 128.74 2487 124.17 74.90 3280 3240 3.83 1520 D50D55D66 67.62 68.03 138.33D100 1709 137.78 68.94DC (b) 1995 136.49 71.56 2357 B 131.97 74.64 3100 124.17 3240 72.13 3.17 2044 D59 68.34 137.31Concordia 2228 Datsin Gangotri 70.00 75.11 12.00 123.32 3232 20 28 135 74 36 13 0.44 2 Table 1 ( M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137–158 143 ) 1 Ϫ yr 2 Ϫ Pb Ref. ux deposition fl 210 Pu ) 2 241 Ϫ Cs 137 ) (a) (b) (Bq m 1 Ϫ yr 2 Ϫ 30.334.336.938.540.1 12342.7 11644.6 5946.3 2447.8 3049.3 4651.3 5338.6 52 44 62 41 66 45 n.m. 68 67 n.m. 32 n.m. 38 n.m. n.m. 66 n.m. n.m. 78 n.m. n.m. 82 n.m. n.m. 68 n.m. n.m. 9, 10 62 n.m. n.m. 9, 10 94 n.m. n.m. 9, 83 10 n.m. n.m. 9, 10 n.m. n.m. 9, 10 n.m. n.m. 9,10 n.m. 9, 10 9, 10 9, 10 9, 10 9, 10 9, 10 17.4 220 134 n.m. n.m. 9, 10 C) (kg m ° Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ Ϫ temperature accumulation Ϫ 8.05 2399 0.462.46 1974 52 120 480 79 56 n.m. n.m. n.m. n.m. 10 10 13.16 362 )(m)( Ϫ Ϫ Ϫ ° Ϫ )( ° ( ) continued Stations Lat. S Long.C E Elevation 10 m 72.26 Snow 2.89 2400 Deposition (Bq m DEFGHIJ 72.51K 72.67L 72.86M 73.04 3.00CV 73.39 3.66CM2 2610 73.72 4.36 2751 74.04 5.04 74.36 2840 6.46 2929 74.64 7.94 74.99 3074 76.00 9.49 11.10 76.10 3174 12.79 3268 3341 15.02 3406 3453 HIKGPS1GPS2M2D2B 70.50 D2 71.51 74.84 70.76 74.64 160.82 74.80 1.18 75.55 157.50 0 1192 75.62 600 1776 151.27 145.79 2308 50 140.63 2454 2611 48 60 53 220 51 69 39 24 27 44 n.m. 56 48 n.m. n.m. 1.10 n.m. 1.72 44 n.m. n.m. n.m. 2.10 n.m. n.m. 1.57 n.m. 2 10 2 10 2 2 2 NARE EF 72.98 1.12 73.10 1817 50 46 n.m. n.m. 10 Table 1 ( 144 M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137–158 Koide et = ) 1 Ϫ yr Pu, and 1965 for 2 241 Ϫ ), and application of 1 Nijampurkar and Rao Ϫ = Pb Ref. ux deposition ;7 fl 210 Pu ) 2 241 Ϫ . 1.6. n.m., not measured. Ref. 1 = 96 41.7 5.095 11 241 107 10.68 11 173.8 79381.66 56 10.10 8.3 11 11 285 52 8.3 11 348.33 94 4.5 11 Cs Cs measurements. We have also to distinguish: 137 Lambert et al. (1983) 137 = Sr/ Cs 90 ;6 137 c activities measurements (Bq kg Roos et al. (1994) fi = ) (a) (b) (Bq m 1 Ϫ yr 2 Pb speci Sanak (1983) Ϫ = 210 ) using the ratio ;5 ∗∗ C) (kg m private com. (cf Isaksson); 11 ° = temperature accumulation ux is calculated from Sr measurements ( fl Pourchet et al. (1997) 90 ;10 = Pb ;4 210 ). ∗ Cs deduced from )(m)( ° 137 Isaksson et al. (1999) = Crozaz (1967, 1969) ux calculations ( = fl ;9 Pb )( 210 ° ( ) this work; 3 Pu are corrected for decay to the deposition time. Deposition times are respectively centered around 1957 for = Cs values, the (a) data series gives the prior measurements and (b) data series, the new 2) 6) 5) 5) Bendezu (1978) 241 = = 137 ;2 = = continued = n n n n ;8 3) Cs and Cs. For Cs deduced from = n Stations Lat. S Long.D4B ED6D ElevationP8 10 mP10P11P12 75.60Lichens 75.45 (mean values) K.G.Is. Snow ( 135.83 77.20 76.64 129.81 2792 76.88 3024 155.50 77.54 147.36 Deposition 141.92 (Bq 2492 m 2477 137.02 2626 2640 21 22 2 2.5 1.6 1.7 40 53 n.m. 28 16 n.m. n.m. 24 1.18 28 n.m. n.m. n.m. 0.36 0.57 n.m. 0.58 0.74 2 2 2 2 2 2 Horsehoe Living ( Mosses, grass, soils (meanK.G.Is. values) ( Living ( Sediment (mean values) ( Table 1 ( The 10-m temperature137 corresponds to the137 value generally used137 to estimate the mean annual temperature for selected area. Global deposition of relations (3) and (6) (see text). al. (1979) (1993) M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137–158 145

Table 2 Distribution of 137Cs and 241Pu ﬂux deposition (Bq mϪ2), and 210Pb speciﬁc activities (Bq kgϪ1) in ice core samples from Vostok and Doˆme Concordia Stations

No. Mean depth 137Cs 241Pu 210Pb (cm) (Bq mϪ2) (Bq mϪ2) (Bq kgϪ1)

Vostok 1 5 0 0 0.0682 ± 0.0136 2 15 0 0 0.1158 ± 0.0232 3 25 0 0 0.0836 ± 0.0168 4 35 0 0 0.1074 ± 0.0214 5 45 0 0 0.0543 ± 0.0108 6 55 0 0 0.0463 ± 0.0092 7 65 0 0 0.0439 ± 0.0088 8 75 0 0 0.0824 ± 0.0164 9 85 0.14 ± 0.03 0 0.0626 ± 0.0126 10 95 0.12 ± 0.02 0 0.0378 ± 0.0076 11 105 0.21 ± 0.04 0 0.0532 ± 0.0106 12 115 0.27 ± 0.05 0 0.0372 ± 0.0074 13 125 0.21 ± 0.04 0 0.0560 ± 0.0112 14 135 0.23 ± 0.05 0 0.0420 ± 0.0084 15 145 0.87 ± 0.17 0 0.0443 ± 0.0088 16 155 2.05 ± 0.41 0 0.0348 ± 0.0070 17 165 3.28 ± 0.66 0 0.0388 ± 0.0078 18 175 5.19 ± 1.04 0 0.0275 ± 0.0054 19 185 9.43 ± 1.88 1.92 ± 0.96 0.0293 ± 0.0058 20 195 11.87 ± 2.37 2.16 ± 1.08 0.0213 ± 0.0042 21 205 15.15 ± 3.03 2.55 ± 1.23 0.0299 ± 0.0060 22 215 12.59 ± 2.52 4.31 ± 2.15 0.0389 ± 0.0078 23 225 6.06 ± 1.12 1.50 ± 0.75 0.0300 ± 0.0060 24 235 5.69 ± 1.14 1.35 ± 0.67 0.0442 ± 0.0088 25 245 11.14 ± 2.23 5.61 ± 2.80 0.0254 ± 0.0050 26 255 18.40 ± 3.68 20.73 ± 10.37 0.0215 ± 0.0042 27 265 5.32 ± 1.06 7.59 ± 3.79 0.0290 ± 0.0058 28 275 0.56 ± 0.11 0.78 ± 0.39 0.0309 ± 0.0062 Doˆme Concordia 1+2 10 0 0 0.0170 ± 0.0034 3+4300 0 0 5450 0 0 6+7 60 0 0 0.0362 ± 0.0072 8 75 0.010 ± 0.002 0 0.0223 ± 0.0045 9850 0 0 10 95 0 0 0.0129 ± 0.0026 11+12 110 0 0 0.0216 ± 0.0043 13 125 0 0 0.0148 ± 0.0030 14 135 0.020 ± 0.004 0 0.0245 ± 0.0049 15 145 0 0 0.0214 ± 0.0043 16 155 0 0 0.0153 ± 0.0031 17 165 0 0 0.0169 ± 0.0034 18+19 180 0.10 ± 0.02 0 0.0107 ± 0.0021 (continued on next page) 146 M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137–158

Table 2 (continued)

No. Mean depth 137Cs 241Pu 210Pb (cm) (Bq mϪ2) (Bq mϪ2) (Bq kgϪ1)

20 195 0.07 ± 0.01 0 0.0116 ± 0.0023 21 205 0.19 ± 0.04 0 0.0133 ± 0.0027 22 215 0.11 ± 0.02 0 0.0131 ± 0.0026 23 225 0 0 0 24 235 0.24 ± 0.05 0 0 25 245 0.39 ± 0.08 0 0 26 255 0.24 ± 0.05 0 0.0094 ± 0.0019 27 265 0.69 ± 0.14 0 0.0124 ± 0.0025 28 275 1.53 ± 0.31 0 0.0183 ± 0.0037 29 285 2.45 ± 0.49 0 0.0069 ± 0.0014 30 295 2.09 ± 0.42 0 0.0151 ± 0.0030 31 305 0.78 ± 0.16 0 0.0154 ± 0.0031 32 315 1.45 ± 0.29 0 0 33 325 5.81 ± 1.16 0 0 34 335 3.99 ± 0.79 0 0.0100 ± 0.0020 35 345 2.08 ± 0.41 0 0.0167 ± 0.0033 36 355 1.91 ± 0.38 0 0.0182 ± 0.0036 37 365 1.59 ± 0.32 0 0.0165 ± 0.0033 38 375 1.27 ± 0.25 0 0 39 385 1.61 ± 0.30 2.21 ± 1.10 0.0079 ± 0.0016 40 395 6.49 ± 1.30 10.91 ± 5.45 0.0240 ± 0.0048 41 405 0.75 ± 0.15 0 0.0195 ± 0.0039 42 415 0 0 0.0099 ± 0.0020 43 425 0 0 0.0097 ± 0.0019

137Cs and 241Pu flux values are decay-corrected to the deposition time (1957 for 241Pu; 1965 for 137Cs). 210Pb concentrations correspond to direct values, at the time measurements (respectively 1999 and 2001). Because of 210Pb half-life (22.3 yrs), there is no decay-correction influence on measured specific activities between 1999 and 2001 values.

corresponding to the arrival and deposition of artiﬁcial radionuclides in Antarctica. 241Pu, on the other hand, is prominent mainly in the older reference level (1955). Indeed, the Yvy and Castles series of tests conducted in 1953 produced more 241Pu than the atmospheric thermonuclear tests in the sixties. This radionuclide therefore makes it possible to distinguish between the two test series. The activity of 137Cs was directly estimated and calculated from high resolution gamma spectrometry results. To estimate the activity of the 241Pu radionuclide (pure alpha emitter), characterized by a short half-life (14.4 yrs), we used the gamma emission of its daughter 241Am, presenting a long half-life (432.7 yrs). 241Pu activity is calculated from the 241Am measurement using a zero value for the daughter at the time of explosion (t0). At time t, the growth of 241Am is calculated using the following equation: 241 ϭ 241 Ϫl1∗tϪ Ϫl2∗t Ϫ Am(t) Pu(t0)[e e ]/[( l2 /l1) 1] (1) M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137–158 147

Fig. 1. 137Cs new density activities versus 137Cs previous data (Pourchet et al., 1997) measured in snow and ice samples for 28 ice-coring stations. Density activities, decay-corrected to deposition time, are expressed in Bq mϪ2. Accuracies for 137Cs are about 20% (see error bars).

Ϫ1 Ϫ1 241 where l1 (0.069 yr ) and l2 (0.00231 yr ) are the radioactive constants for Pu and 241Am respectively. Based on the radioactive decay law, the maximum activity of 241Am occurs 32 yrs after 241Pu production: ϭ Ϫ tm [1/(l2 l1)]log10(l2 /l1) (2)

241 where tm is time of maximum activity of the daughter radionuclide, Am in this case. tm is expressed in yrs. In agreement with detailed proﬁles of radionuclide deposition on the Ross Ice Shelf (Koide et al., 1979) and at Vostok and Dome Concordia (this work), and also with the well documented arrival dates of total beta peaks (Jouzel et al., 1979), we have assumed that 137Cs and 241Pu deposition are respectively centered around 1965 and 1957. In fact, we have to clearly distinguish between the time of maximum activity (1955 and/or 1965), well-known in radioactive fallout over Antarctica, and the time of the mean deposition of each radionuclide 137Cs and 241Pu. Indeed, the 241Pu proﬁle in one ice-core sample showed a maximum activity in 1955, but the mean value of the total deposition is considered to be centered around 1957. For 137Cs, the maximum activity and the mean of the total deposition occur simul- taneously, i.e. in 1965. Tables 1 and 2 summarize 137Cs, 210Pb and 241Pu activities (expressed in Bq mϪ2) found in the different samples. Note that the absence of activity values for the 241Pu nuclide in some cases does not indicate the total absence 148 M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137–158 of this radionuclide in the snow. In these cases, the activity of 241Pu is simply so low that it cannot be distinguished due to the detection limit of the analyzer for the counting times used.

2.4. Measurements and determination of 210Pb

Lead 210, a natural beta emitter, is a long-lived daughter nuclide (half-life 22.3 yrs) belonging to the Uranium 238 family (Picciotto and Crozaz, 1971). Its presence in the atmosphere is a result of the alpha radioactive decay of radon gas (222Rn). Note that the atmosphere, via tropospheric and stratospheric circulation, is the major source of 210Pb in Antarctica. The permanent ice over the Antarctic continent that prevents the escape of radon from geological substrates, the surrounding ocean with- out radon emanation and the time required for air masses to move from continental areas (source of radon emission) to south polar region all result in a low radon (and its daughter nuclide) concentration in Antarctica (Pourchet et al., 1997). In this work, 210Pb was directly analyzed by gamma spectrometry. The deposition of 210Pb and the related natural fluxes are summarized in Tables 1 and 2. 210 Ϫ2 The mean value of the continuous Pb deposition flux (A0 expressed in Bq m yrϪ1) can be estimated by: ϭ Ϫ Ϫ Ϫ A0 lS/[exp( lt1) exp( lt2)] (3) where l is the 210Pb radioactive decay constant (l = 0.03114 yrϪ1); S is the remaining activity of 210Pb in the total core at the time of measurement, using the mean annual Ϫ2 accumulation rate; this value is expressed in Bq m ; t1 is the elapsed time (expressed 210 in yrs) between core sampling and Pb measurement; t2 is the elapsed time (expressed in yrs) between deposition of the deepest part of the core analyzed and the 210Pb measurement. In this work, in order to estimate 210Pb deposition flux, we have systematically chosen to analyze data between the first reference layer of 1955 and the top of the core (or sometimes a few meters below the top). This interval was previously the subject of total beta radioactivity measurements used to determine the mean annual accumulation.

3. Results

3.1. Comparison between prior and new 137Cs determinations

As already pointed out, we carried out new 137Cs determinations for 28 previously analyzed stations (Pourchet et al., 1997). For all 28 stations, old and new 137Cs measurements are linked by the following equation: Y (prior 137Cs values) ϭ 0.8615 X (new 137Cs values) R2 ϭ 0.91 (4) (n ϭ 28) M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137–158 149 where prior and new 137Cs values are expressed in Bq mϪ2. However, a detailed analysis of the scatter plot of old 137Cs vs. new 137Cs shows two outlying pairs of measurements. After eliminating these pairs, the relation between old 137Cs vs. new 137Cs shows a linear correlation of 0.98. The relation is as follows: Y (prior 137Cs values) ϭ 0.9225 X (new 137Cs values) R2 ϭ 0.98 (5) (n ϭ 26) This trend conﬁrms the validity of previous results (Pourchet et al., 1997) and that of the analytical procedure.

3.2. History and Vostok and Dome Concordia chronologies

The chronology of the arrival of undifferentiated or major artificial radionuclides is very well known in Antarctica. The first significant increase (peak intensity) of total beta radioactivity was observed in January 1955 (Picciotto and Wilgain, 1963; Vickers, 1963; Woodward, 1964; Wilgain et al., 1965) and corresponds to the arrival of the Yvy and Castles atmospheric thermonuclear tests, conducted in the Pacific Islands in November 1952 and March 1954, respectively. The continuation of high- yield bomb tests in the atmosphere, despite the nuclear moratorium in 1960, main- tained the radioactivity of Antarctic precipitation at a significant level. The minimum fallout observed in 1961, was then followed by an increase reaching a maximum in 1964–1965 (Picciotto and Crozaz, 1971). The highest recorded total activity due to the American and U.S.S.R. test series arrived in January 1965 (Wilgain et al., 1965; Lambert et al., 1977; Jouzel et al., 1979). Note that the marked activity surge (highest intensity peak) in the 1955 reference layer with regard to total activity in 1964–1965 is due to a combination of several factors including the very high power of first detonation of the Castle test series, the southern hemisphere location of the test and the time of the year at which it was set off (rapid transfer into the atmosphere and to the ground). The two reference layers were clearly identified in the Vostok 137Cs profile at respectively 250–260 cm (88.3–92.0 cm w.e.) and 200–210 cm (70.1–73.7 cm w.e.) depths (Fig. 2a) and should be considered as two unambiguous references for dating 210Pb and 241Pu deposition. As we have already pointed out, 241Pu deposition in Vostok is mainly concentrated in the 1955 layer. This is in accordance with Ross Ice Shelf 241Am deposition (Koide et al., 1979), the 239+240Pu snow profile at Dome C(Cutter et al., 1979) and our results at Dome Concordia (Fig. 2b).

3.3. 210Pb dating and origin

The mean annual accumulation for the 1955–1998 and 1965–1998 periods deduced from both reference levels are respectively 2.05 and 2.18 cm of water equivalent (w.e.). For the 1955–1998 period, the mean annual accumulation deduced from the radioactive decay of 210Pb in the snow proﬁle is 2.57 cm w.e. The conversion of 150 M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137–158

Fig. 2. 137Cs, 241Pu and 210Pb profiles versus depth of two shallow Vostok (a) and Doˆme Concordia (b) ice cores, with the 1955 and 1965 radioactive reference layers (indicated by an arrow). Radionuclides deposition (137Cs and 241Pu) are expressed in Bq m–2, and 210Pb specific activity, in Bq kgϪ1. Depth is expressed in cm of snow equivalent. M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137–158 151 real snow depths to water equivalents eliminates all distortion related to differences in the compaction of snow layers. The activity density profile of 210Pb (Bq kgϪ1)is therefore expressed as a function of the depth expressed in cm of water equivalent. For this estimation, we assumed a constant deposition flux of 210Pb from the atmosphere to the snow, a closed system for lead following its deposition, and no changes in 210Pb activity through exchanges by diffusion or percolation of melt water (Picciotto and Crozaz, 1971). This value is significantly higher than the absolute 137Cs determination. Possible reasons for this discordance include the following factors:

(i) a change in the annual 210Pb deposition flux; (ii) a change in the annual snow accumulation; 210 210 210 (iii) part of the total Pb ( Pbt) content in the snow coming from Pb supported 238 210 by long-life radionuclides of the U family ( Pbs). The first reason can be eliminated given the known constant 210Pb deposition flux over the time period in question. Similarly, the second reason can be eliminated given the absolute dating provided on the basis of the 1955 and 1965 reference layers with respect to the surface.For the last reason, the 210Pb dating, in all snow samples, 210 210 Ϫ1 must be calculated using Pb excess ( Pbe) expressed in Bq kg from: 210 ϭ 210 Ϫ210 Pbe Pbt Pbs (6) 210 210 where Pbe is the excess or unsupported Pb activity produced in the atmosphere 222 210 210 by radioactive decay of Rn, Pbt the total Pb activity measured in snow 210 210 samples and Pbs the Pb activity supported by long-life radionuclides of the 238U family. For the 1955–1998 period a similar value of the yearly average accumulation deduced from 137Cs or 210Pb is obtained when we apply a constant value of 9 mBq Ϫ1 210 210 210 kg to the Pbs. With this value, Pbe and Pbs are equal for a 65-year old snow sample. This result is consistent with a previous estimation based on 226Ra 210 measurements at Dome C, where the same activity for both parts of Pbt was calculated for 60-year old snow (Pourchet et al., 1997). Note however that snow redistri- bution by wind should not be neglected as it can disturb snow and 210Pb deposition over several yrs, and consequently the 210Pb dating. At Dome Concordia, the 137Cs and 241Pu profiles (Fig. 2b) are very similar to the Vostok profiles and give a mean annual accumulation of 3.2 cm w.e. since 1955. Even if the accuracy of the 241Pu values is only about 50%, we obtain accurate mean annual accumulation values because the depth of corresponding reference layer is very precise. However, the 210Pb profile is very disturbed and does not allow proper 210 dating or an estimation of the unsupported Pbe. 3.4. Total deposition and fluxes of radionuclides

3.4.1. Radionuclide deposition and the 137Cs/241Pu relationship For the whole studied deposition period (1955–1980), the total 137Cs fallout over all the stations analyzed (Table 1) varies from 16 (P8 station) to 181 Bq mϪ2 (km 152 M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137Ð158

200), with a mean value of 64 ± 32 Bq mϪ2. The median of 56 Bq mϪ2 lies 12.5% below the mean value, revealing an asymmetric distribution of 137Cs, as observed in Pourchet et al. (1997). Some very high values from 200 km, R. Baudoin, M.M., DB, D45, D53, New Byrd and CM2 stations can be considered as outliers. A map of the deposition and distribution of 137Cs over Antarctica for the whole deposition period (1955–1980) shows that radionuclide fallout is higher over coastal areas. We deliberately decided to describe the radionuclide distribution in the most documented and studied area in Antarctica, i.e. the 90–180° East sector (Fig. 3). Data from the other sectors of Antarctica cover shorter periods and would likely be insufficient to draw conclusions on a global radionuclide distribution law. Despite major scatter in individual values, classification of this data into several groups of accumulation rates clearly shows a positive correlation between 137Cs deposition and accumulation. As well, we observe a discontinuity in the relation (Fig. 4) near an accumulation rate of 150 kg/m2/year, indicating the possible existence of two radionuclide distribution laws. This discontinuity, already observed (Pourchet et al., 1997), lies between coastal areas and the polar plateau. The first classes, characterized by low accumulation rates and 137Cs fall-out, correspond to the plateau area. The second group includes the sampling sites near the coast. Note that the leaching of radioactive aerosols is higher for sampling sites characterized by low accumulation rates (e.g. Antarctic plateau). Assuming that is possible to extrapolate 137Cs flux values to zero accumulation, a “dry flux” of 43 Bq mϪ2 is obtained for 137Cs, which is consistent with a previous estimation of “dry deposition” (Pourchet et al., 1997). Comparison of 137Cs distribution in three latitudinal classes shows discontinuities in deposition. For 60– 70°,70–80° and 80–90° lat. S, mean 137Cs deposition is respectively 77, 58 and 57

Fig. 3. Deposition and distribution Antarctic map 137Cs (sector 90–180° East) for the whole deposition period 1955–1980. M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137Ð158 153

Fig. 4. Variation of total 137Cs deposition versus accumulation rates in sector 90–180° East of Antarctica, and illustration of existence of two radionuclide distribution laws between coastal areas and polar plateau. Density activities values represent the 1955–2000 period. Radionuclide deposition is expressed in Bq mϪ2, and accumulation rates in kg mϪ2 yearϪ1.

Bq mϪ2. A clear change in 137Cs deposition processes is observed close to the break- ing slope on Antarctic plateau, near 80° lat. S. The maximum deposition of 241Pu is observed at 200 km, but many stations (57% of stations studies) are below the detection limit (between 5 and 10 Bq mϪ2), depending on the counting time. The total fallout of 241Pu over stations for which this radionuclide is detectable (Table 1), varies from 9 (F9 station) to 99 Bq mϪ2 (200 km), with a mean value of 37 ± 20 Bq mϪ2. Regarding the 137Cs distribution, the median value of 241Pu, i.e. 36 Bq mϪ2, is close to the mean value, demonstrating the symmetry of the distribution of this radionuclide. Nevertheless, given the large dispersion of values around the mean and the low number of results studied, we must be cautious when analyzing 241Pu. Note also that the least squares method was used for statistical analysis and the linear best-ﬁt regression was forced through the origin. Considering only the stations for which 241Pu is clearly detectable, i.e. 28 stations, 137Cs and 241Pu total deposition values are highly correlated (Fig. 5a) according to the following equation: [241Pu ] ϭ 0.484 [137Cs ] R 2 ϭ 0.92 (n ϭ 28) (7) This relationship is very close to that obtained from lichens (Alectoria nigricans 154 M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137Ð158

Fig. 5. 241Pu versus 137Cs deposition (a) and 210Pb flux versus 137Cs deposition (b) in different types of samples (snow, soils and lichens). Radionuclides deposition are both expressed in Bq mϪ2, and 210Pb flux deposition in Bq mϪ2 yrϪ1. M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137Ð158 155 species) collected on the Antarctic Peninsula area by Roos et al. (1994) (eight sampling stations): [241Pu ] ϭ 0.454 [137Cs ] R 2 ϭ 0.99 (n ϭ 8) (8) At the time of deposition, the mean 137Cs/241Pu ratio for total deposition is 1.9 for the snow stations and 2.3 for the lichens (Table 3). The detailed profiles of Vostok and Dome Concordia give similar values (2.2 and 2.7, respectively). This relative concordance between 137Cs and 241Pu is not observed on various types of moss and grass, sampled during the SWEDARP expedition, because of the effects of submergence and meltwater that alter the radionuclide ratios, as explained by Roos et al. (1994). Similarly, sediment samples collected on the Antarctic Peninsula show strong variations in the 137Cs/241Pu ratio and no relationship has been found. Normally sediments, like snow, are good accumulators of tracers (radionuclides in our case) and therefore could provide a 137Cs/241Pu ratio close to that observed in snow and lichen samples. However, the very low rate of sedimentation in the lakes concerned (30–100 cm to 1000 yrϪ1) and the large surrounding snow fields delivering meltwater to the lakes explain the high variations (Roos et al., 1994).

3.4.2. Reference levels for the 137Cs/241Pu at Vostok and DC The very large Pu ratios (137Cs/241Pu) observed during the early US test series are conﬁrmed for Vostok station by the separate 1955 and 1965 peak ratios (respectively 0.9 and 5.9). Similar trends have previously been observed for the Ross Ice Shelf (Koide et al., 1979) but with different absolutes values. The 1965 value measured at Dome Concordia is too low for a signiﬁcant evaluation of this ratio. With 241Pu deposition centered at 1957, the maximum activity of 241Am at the same level would occur 32 yrs later, i.e. in 1989. Due to its long half life, the relative importance of this radionuclide is increasing.

Table 3 Spatial variations of 137CsϪ241Pu ratio in different Antarctic locations. and comparison of the calculated ﬁssion products relationships in different types of samples (snow, sediments, soils and lichens)

137Cs/ 241Pu Snow Vostok Dome J9 (1) South Shetland Island (11) stations Concordia

lichens moss, grass, sediments soils

1955 peak 0.9 0.6 1.6 1965 peak 5.9 9.2 Mean value 1.9 2.2 2.7 5.4 2.3 7.1 5.9 (n=27) (n=8) (n=11) (n=3)

The number of determinations (n) used to calculate the mean value are indicated in brackets. The biblio- graphic references for the station J9 (1) and the South Shetland Island (11) correspond respectively to Koide et al. (1979) and Roos et al. (1994) (see Table 1). 156 M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137Ð158

3.4.3. Mean 210Pb/137Cs ratio and relationship 210 210 137 The Pb deposition ﬂux ( Pbf) is linked to total Cs deposition (Fig. 5b)by the relation: 210 ϭ 137 2 ϭ ϭ [ Pbf] 0.041 [ Cs] R 0.84 (n 49) (9) Compared to the 137Cs/241Pu relation, the 210Pb/137Cs ratios are very close for snow or ice-core samples and the lichens studied by Roos et al. (1994). The good correlation between the different radionuclides suggests the creation of an undifferentiated radionuclide deposition (or ﬂux) map over Antarctica. With this aim, we have calculated the equivalent 137Cs deposition (Table 1) using all available 210Pb or 90Sr data.

4. Discussion

The proposed map is constructed using a constant ratio between the different radionuclides. This relationship does not take into account the differences in the fallout mechanisms for radionuclides observed elsewhere, for example the differences in the dry and wet deposition between 210Pb, 90Sr and 137Cs (Pourchet et al., 1997). In this work we have estimated the 210Pb deposition ﬂux from total 210Pb and not 210 210 from the direct atmospheric Pb ( Pbe). The two values differ by 6%. For 241Pu estimates based on 241Am measurements, we have assumed a zero value of 241Am at the time of deposition. In fact the real zero value is the detonation time about 1.5 year earlier. The resulting error is less than 10%. The greatest source of dispersion between measured and estimated values is the lack of representativeness of the snow samples. Local topography (dunes, sastrugis) and drifting of the snow are very present in Antarctica. For 21 stations, the 137Cs budget is less than the estimated dry deposition, revealing local partial erosion of the deposited snow. At Vostok station, the roughness of the snow surface is equivalent to ﬁve yrs of deposition (30 cm), centered around 1965 and representing 40% of the 137Cs budget. This could explain the two different 137Cs budgets observed at this location. Due to the very short duration of 241Pu deposition, this radionuclide will be more affected by this phenomenon. This could explain the differences between 241Pu deposition at several stations. The irregular geographical distribution of the measured points does not allow us to establish reliable equidensity scatter plots. For example, to validate the estimated deposition values between Terra Nova and Dumont d’Urville, more samples are necessary (Fig. 3). The mean deposition of 137Cs in Antarctica obtained from data in the literature is clearly lower than the values estimated (240–260 Bq mϪ2) from soils sampled on South Shetland Island (60° lat. S) by Roos et al. (1994) and Godoy et al. (1998). Similarly, deposition values published by UNSCEAR (1982) indicate 300, 180 and 70 Bq mϪ2 for latitudes from 60 to 70° S, 70 to 80° S and 80 to 90° S, respectively. Previous values seem to overestimate 137Cs deposition in Antarctica. The limited M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137Ð158 157 number of samples and related values used to calculate these deposition values could be the main reason for the difference between their estimations and our results. With an area of 13,500,000 km2 and a median value of 56 Bq mϪ2, the total deposition of 137Cs over Antarctica is estimated at 760 TBq, representing less than 0.08% of the total worldwide deposition of this radionuclide. 137Cs and 210Pb transported in the stratosphere represent excellent atmospheric tracers and could be used to estimate the deposition of other sources of pollution (heavy metals, pesticide, etc.) found in Antarctica.

Acknowledgements

The authors are grateful to all colleagues who participated in ﬁeld work and sampling operations. Thanks are also due to C. Vincent (LGGE) for his help in the preparation of this paper and to J. Ras (Marine station, LPCM), and Harvey Harder for her help in improving the English.

References

Bendezu, A.M., 1978. Application des mesures de plomb 210 et Sr-90 a` l’e´tude des ećhanges entre l’atmosphe`re et la calotte Antarctique. PhD thesis, Universite´ de Paris. Crozaz, G., 1967. Mise au point d’une me´thode de datation des glaciers baseé sur la radioactivite´ du Pb- 210. The`se de doctorat, Universite´ libre de Bruxelles. Crozaz, G., 1969. Fission products in antarctic snow: an additional reference level in January 1965. Earth Planet. Sci. Let., 6 (1), 6–8. Cutter, G.A., Bruland, K.W., Risebrough, R.W., 1979. Deposition and accumulation of plutonium isotopes in Antarctica. Nature 279, 628–629. Delmas, R., Pourchet, M., 1977. Utilisation des filtres ećhangeurs d’ions pour l’e´tude de l’activite´ β globale d’un carottage glaciologique. IAHS Publ 118, 159–163. Feely, H.W., Seitz, H., Lagomarsino, R.J., Biscaye, P.E., 1966. Transport and fallout of stratospheric radioactive debris. Tellus 18 (2–3), 316–328. Godoy, J.M., Schuch, L.A., Nordemann, D.J.R., Reis, V.R.G., Ramalho, M., Recio, J.C., Brito, R.R.A., Olech, M.A., 1998. 137Cs, 226,228Ra, 210Pb and 40K concentrations in Antarctic soils, sediment and selected moss and lichen samples. J. Environ. Radioactivity 41 (1), 33–45. Isaksson, E., Van Den Broeke, M.R., Winther, J.G., Karlof, L., Pinglot, J.F., Gundestrup, N., 1999. Accumulation and proxy-temperature variability in Dronning Maud Land, Antarctica, determined from shallow firn cores. Annals of Glaciology 29, 17–22. Jouzel, J., Merlivat, L., Pourchet, M., Lorius, C., 1979. A continuous record of artificial tritium fallout at the south pole. Earth Planet. Sci. Lett. 45, 188–200. Koide, M., Goldberg, E.D., Herron, M.M., Langway, C.C., 1979. Deposition history of artificial radionuclides in the Ross Ice Shelf, Antarctica. Earth Planet. Sci. Lett. 44, 205–223. Lambert, G., Ardouin, B., Lorius, C., Pourchet, M., 1977. Accumulation of snow and radioactive debris in Antarctica: a possible refined radiochronology beyond reference levels. IAHS Publ. 118, 146–158. Lambert, G., Ardouin, B., Mesbah-Bendezu, A., 1983. Atmosphe`re to snow transfert in antarctica. In: Pruppacher, H.R., Semonin, R.G., Slinn, W.G.N. (Eds.), Precipitation Scavenging, Dry Deposition, and Resuspension. New York, Elsevier Science Publishers, pp. 1289–1300. Nijampurkar, V.N., Rao, D.K., 1993. Polar fallout radionuclides 32Si, 7Be and 210Pb and past accumulation rate of ice at Indian Station, Dakshin Gangotri, East antarctica. J. Environ. Radioactivity 21, 107–117. 158 M. Pourchet et al. / J. Environ. Radioactivity 68 (2003) 137Ð158

Picciotto, E., Wilgain, S., 1963. Fission products in Antarctic snow, a reference level for measuring accumulation. J. Geophys. Res. 68 (21), 5965–5972. Picciotto, E., Crozaz, G., 1971. Accumulation on the south pole-queen Maud Land traverse, 1964–1968. Antarctic Snow and Ice studies II in Antarctic Research Series 16, 257–315. Pinglot, J.F., Pourchet, M., 1979. Low level Beta counting with an automatic sample changer. Nuclear Instruments and Methods 166, 483–490. Pinglot, J.F., Pourchet, M., 1994. Spectrome´trie gamma a` tre`s bas niveau avec anti- Compton NaI (TI) pour l’e´tude des glaciers et des se´diments. Note CEA-N-2756, ISSN 0429-3460. Pinglot, J.F., Pourchet, M., 1995. Radioactivity measurements applied to glaciers and lake sediments. Sci. Total Environ. 173/174, 211–223. Pinglot, J.F., Pourchet, M., Lefauconnier, B., Hagen, J.O., Isaksson, E., Vaikmac, R., Lamiyama, K., 1999. Accumulation in Svalbard glaciers deduced from ice cores with nuclear tests and Chernobyl reference layers. Polar Research 18 (2), 315–321. Pourchet, M., Pinglot, J.F., Lorius, C., 1983. Some meteorological applications of radioactive fallout measurements in Antarctic snows. J. Geophys. Res. 88 (C10), 6013–6020. Pourchet, M., Bartaya, S.K., Maignan, M., Jouzel, J., Pinglot, J.F., Aristarain, A.J., Furdada, G., Kot- lyakov, V.M., Mosley-Thompson, E., Preiss, N., Young, N.W., 1997. Distribution and fall-out of 137Cs and other radionuclides over Antarctica. J. Glaciology 43 (45), 435–445. Roos, P., Holm, E., Persson, R.B.R., Aarkrog, A., Nielsen, S.P., 1994. Deposition of 210Pb, 137Cs, 239+240Pu, 238Pu, and 241Am in the Antarctic Peninsula area. J. Environ. Radioactivity 24, 235–251. Sanak, J., 1983. Contribution a` l’e´tude du transport des ae´rosols d’origine continentale vers l’antarctique. The`se de doctorat, Universite´ pierre et Marie Curie, Paris. United Nations Scientiﬁc Committee on the effects of Atomic Radiation, UNSCEAR, 1982. Sources, effects and risks of ionizing radiation. United Nations, New York. Vickers, W.W., 1963. Geochemical dating techniques applied to Antarctic snow samples. General Assembly, Berkeley, August 1963. Intern. Assoc. Sci. Hydrol. Publ. 61, 199–215. Wilgain, S., Picciotto, E., de Breuck, W., 1965. Strontiumout 90 fallout in Antarctica. J. Geophys. Res. 70 (24), 6023–6032. Woodward, R.N., 1964. Strontiun 90 and Caesium 137 in Antarctic snows. Nature 204, 1291.